11 accepted papers
Post-hoc interpretability methods typically attribute a model’s behavior to its components, data, or training trajectory in isolation. This leads to explanations that lack a unified view and may miss key interactions. While combining existing methods or applying them at different training stages off…
Transformers have shown inconsistent success in AI planning tasks, and theoretical understanding of when generalization should be expected has been limited. We take important steps towards addressing this gap by analyzing the ability of decoder-only models to verify whether a given plan correctly so…
This paper introduces the TempVS benchmark, which focuses on temporal grounding and reasoning capabilities of Multimodal Large Language Models (MLLMs) in image sequences. TempVS consists of three main tests (i.e., event relation inference, sentence ordering and image ordering), each accompanied with…
Data pruning—selecting small but impactful subsets—offers a promising way to efficiently scale NLP model training. However, existing methods often involve many different design choices, which have not been systematically studied. This limits future developments. In this work, we decompose data pruni…
Despite the massive success of fine-tuning Pre-trained Language Models (PLMs), they remain susceptible to out-of-distribution input. Dataset cartography is a simple yet effective dual-model approach that improves the robustness of fine-tuned PLMs. It involves fine-tuning a model on the original trai…
Implicit reasoning is the ability of a language model to solve multi-hop reasoning tasks in a single forward pass, without chain of thought.We investigate this capability using GPT2-style language models trained from scratch on controlled k -hop reasoning datasets ( k = 2, 3, 4 ). We show that while
Language models excel in various tasks by making complex decisions, yet understanding the rationale behind these decisions remains a challenge. This paper investigates data-centric interpretability in language models, focusing on the next-word prediction task. Using representer theorem, we identify…
Large language models readily memorize arbitrary training instances, such as label noise, yet they perform strikingly well on reasoning tasks. In this work, we investigate how language models memorize label noise, and why such memorization in many cases does not heavily affect generalizable reasonin
Fine-tuning pre-trained language models on downstream tasks with varying random seeds has been shown to be unstable, especially on small datasets. Many previous studies have investigated this instability and proposed methods to mitigate it. However, most of these studies only used the standard devia…
Knowledge base (KB) embeddings have been shown to contain gender biases. In this paper, we study two questions regarding these biases: how to quantify them, and how to trace their origins in KB? Specifically, first, we develop two novel bias measures respectively for a group of person entities and a…
Various measures have been proposed to quantify human-like social biases in word embeddings. However, bias scores based on these measures can suffer from measurement error. One indication of measurement quality is reliability, concerning the extent to which a measure produces consistent results. In…